Introducing Edna: A trainee chatbot designed to support communication about additional (secondary) genomic findings

Area of Science:

• Genomic Medicine
• Bioinformatics
• Health Informatics

Background:

• Clinical genomic data reanalysis for preventable conditions ('additional findings') requires informed patient decision-making.
• Existing genetic counseling processes can be augmented with digital tools to improve patient support.

Purpose of the Study:

• To develop a chatbot, named electronic genetic resource (eDNA), to support informed decision-making regarding the reanalysis of clinical genomic data for additional findings.
• To enhance patient agency and health literacy concerning genetic testing and potential risks of preventable conditions.

Main Methods:

• Characterization of genetic counseling interactions through thematic and semantic analysis of transcripts.
• Development of interfaces to integrate external genetics application data.
• Programming the chatbot using a framework that supports speech-to-text interaction with patients.

Main Results:

• Chatbot conversations were encoded with conditions, terms, explanations, and decision-making factors, emulating counseling flows.
• Patient agency and health literacy were enhanced through prompted consideration of personal/familial implications and explanations of genetic concepts.
• Novel features included sentiment analysis and family history collection; medical advice was excluded.

Conclusions:

• The successful development of eDNA signifies progress towards accessible and well-supported digital health solutions for patient decision-making on additional findings.
• eDNA complements traditional genetic counseling by providing and collecting genomic information flexibly around consultations.